Determination of vesicle size distributions by freeze-fracture electron microscopy

Spontaneous Formation of Ultrasmall Unilamellar Vesicles in Mixtures of an Amphiphilic Drug and a Phospholipid

We have observed ultrasmall unilamellar vesicles, with diameters of less than 20 nm, in mixtures of the tricyclic antidepressant drug amitriptyline hydrochloride (AMT) and the unsaturated zwitterionic phospholipid 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC) in physiological saline solution. The size and shape of spontaneously formed self-assembled aggregates have been characterized using complementary techniques, i.e., small-angle neutron and X-ray scattering (SANS and SAXS) and cryo-transmission electron microscopy (cryo-TEM). We observe rodlike mixed micelles in more concentrated samples that grow considerably in length upon dilution, and a transition from micelles to vesicles is observed as the concentration approaches the critical micelle concentration of AMT. Unlike the micelles, the spontaneously formed vesicles decrease in size with each step of dilution, and ultrasmall unilamellar vesicles, with diameters as small as about 15 nm, were observed at the lowest concentrations. The spontaneously formed ultrasmall unilamellar vesicles maintain their size for as long we have investigated them (i.e., several months). To the best of our knowledge, such small vesicles have never before been reported to form spontaneously in a biocompatible phospholipid-based system. Most interestingly, the size of the vesicles was observed to be strongly dependent on the chemical structure of the phospholipid, and in mixtures of AMT and the phospholipid 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), the vesicles were observed to be considerably larger in size. The self-assembly behavior in the phospholipid-drug surfactant system in many ways resembles the formation of equilibrium micelles and vesicles in mixed anionic/cationic surfactant systems.

Multilamellarity, structure and hydration of extruded POPC vesicles by SANS

The small-angle neutron scattering (SANS) data of 12 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) dispersions at low lipid concentration (1 mg per 100-mg heavy water) prepared by 5, 9 and 29 extrusions through filters of pores with 50, 100, 200 and 400 nm diameter are presented. They were analyzed within a theory that permits the determination of both structural and hydration parameters of the bilayers as well as the portions of multilamellar vesicles in dispersions with negligible long-range order between the vesicles. The scattering length density profile across the bilayers is approximated by assuming a central hydrocarbon core surrounded by a water-accessible coat. It is modeled by two different forms of functions. In the boat model, the scattering length density of the coat changes linearly from core to water, whereas in the strip model it is constant across the water-accessible coat. It was found that the boat model reflects the reality better than the strip model. The decrease of the multilamellar vesicle portions, either with increasing the number of extrusions at same filter size and with decreasing the filter size, was characterized quantitatively.

Determination of bilayer thickness and lipid surface area in unilamellar dimyristoylphosphatidylcholine vesicles from small-angle neutron scattering curves: a comparison of evaluation methods

Small-angle neutron scattering (SANS) experiments were performed on unilamellar 1,2-dimyristoylphosphatidylcholine (DMPC) vesicles prepared in heavy water by extrusion through polycarbonate filters with 500 A pores. The data obtained at 30+/-0.1 degrees C were evaluated using a five-strip function model of the bilayer coherent neutron scattering length density, three different approximate form factors describing scattering from vesicles, and different methods of evaluation of the experimental data. It is shown that the results obtained from the SANS data in the range of scattering vector values 0.0316 A(-1)< q<0.0775 A(-1) are not sensitive to the vesicle form factor, nor to the evaluation method. Using the hollow sphere model of vesicles convoluted with the Gaussian distribution of their sizes, a constrained bilayer polar region thickness of 9 A and a DMPC headgroup volume of 325.5 A(3), it was possible to obtain from the experimental data the DMPC surface area as 58.9+/-0.8 A(2), the bilayer thickness as 44.5+/-0.3 A and the number of water molecules as 6.8+/-0.2 per DMPC located in the bilayer polar region.

The crossflow injection technique: an improvement of the ethanol injection method

A novel scalable liposome preparation technique for pharmaceutical application is presented. Previous experiments have shown that the concept of continuous crossflow injection is a promising approach. For the characterization of the process, we focus on the influencing parameters like the lipid concentration, the injection hole diameter, the injection pressure, the buffer flow rate, and system performance. These experiments demonstrate that the injection hole diameter and the system performance do not influence the vesicle forming process and that a minimum of buffer flow rate is required to affect batch homogeneity. In contrast, strongly influencing parameters are lipid concentration in combination with increasing injection pressures. After exceeding the upper pressure limit of the linear range, where injection velocities remain constant, the vesicle batches are narrowly distributed, also when injecting higher lipid concentrations. Reproducibility and scalability data show similar results with respect to vesicle size and size distribution and demonstrate the stability and robustness of the novel continuous liposome preparation technique.

Transfection properties of stabilized plasmid-lipid particles containing cationic PEG lipids

Recent work has shown that plasmid DNA can be efficiently encapsulated in well-defined "stabilized plasmid-lipid particles" (SPLP) that have potential as systemic gene therapy vehicles [Gene Ther. 6 (1999) 271]. In this work, we examine the influence of ligands that enhance cellular uptake on the transfection potency of SPLP. The ligand employed is a cationic poly(ethylene glycol) (PEG) lipid (CPL) consisting of a lipid anchor and a PEG(3400) spacer chain with four positive charges at the end of the PEG (CPL(4)). It is shown that up to 4 mol% CPL(4) can be inserted into preformed SPLP, resulting in up to 50-fold enhancements in uptake into baby hamster kidney (BHK) cells. The addition of Ca(2+) to SPLP-CPL(4) (CPL(4)-incorporated SPLP) results in up to 10(6)-fold enhancements in transgene expression, as compared to SPLP in the absence of either CPL(4) or Ca(2+). These transfection levels are comparable to those observed for plasmid DNA-cationic lipid complexes (lipoplexes) but without the cytotoxic effects noted for lipoplex systems. It is concluded that in the presence of Ca(2+) and appropriate ligands to stimulate uptake, SPLP are highly potent transfection agents.

Structure of artificial cytoskeleton containing liposomes in aqueous solution studied by static and dynamic light scattering

The structure of three types of liposomes (egg yolk phosphatidylcholine (EPC) without modification and EPC vesicles containing cross-linked N-isopropylacrylamide (NIPAM) networks of low and a high concentration inside the vesicles) were analyzed by static and dynamic light scattering. Upon polymerization the network was assumed to become attached to the membrane by reactive anchoring monomers. For the sample of high poly(NIPAM) content the polymer network was assumed to fill the whole space in the vesicles. The issue of the present study was to examine hard and hollow sphere behavior of the liposomes with networks of high and low poly(NIPAM) content. The theoretical scattering curves differ markedly for uniform hard and uniform hollow spheres by the presence of specific peaks. However, polydispersity washed out the peaks and led to smoothed asymptotes with fractal dimensions of df = 2 for hollow and df = 4 for hard spheres. The experimental data could efficiently be fitted with weakly polydisperse hollow spheres. No clear conclusion could be drawn from the angular dependence alone for the liposome of high poly(NIPAM) content. The two wavelengths from the HeNe and Ar lasers proved to be too long for the studied liposomes of about 100 nm in radius. However, evidence for hollow sphere behavior was found for fractionated liposomes from the ratio rho = Rg/Rh = 1.04 +/- 0.02 (theory rho = 1.00 for hollow spheres). Finally, from the molar mass and the sphere radius, an apparent density was determined. The analysis gave the expected density for the pure EPC lecithin vesicles and a poly(NIPAM) network density of 0.244 g/mL. For the liposome of low poly(NIPAM) content the network appeared to be attached to the inner surface of the lecithin shell to form a layer of about 18 nm thickness.

Osmotically induced shape changes of large unilamellar vesicles measured by dynamic light scattering

Static and dynamic light scattering measurements have been used to characterize the size, size distribution, and shape of extruded vesicles under isotonic conditions. Dynamic light scattering was then used to characterize osmotically induced shape changes by monitoring changes in the hydrodynamic radius (R(h)) of large unilamellar vesicles (LUVs). These changes are compared to those predicted for several shapes that appear in trajectories through the phase diagram of the area difference elasticity (ADE) model (. Phys. Rev. E. 52:6623-6634). Measurements were performed on dioleoylphosphatidylcholine (DOPC) vesicles using two membrane-impermeant osmolytes (NaCl and sucrose) and a membrane-permeant osmolyte (urea). For all conditions, we were able to produce low-polydispersity, nearly spherical vesicles, which are essential for resolving well-defined volume changes and consequent shape changes. Hyper-osmotic dilutions of DOPC vesicles in urea produced no change in R(h), whereas similar dilutions in NaCl or sucrose caused reductions in vesicle volume resulting in observable changes to R(h). Under conditions similar to those of this study, the ADE model predicts an evolution from spherical to prolate then oblate shapes on increasing volume reduction of LUVs. However, we found that DOPC vesicles became oblate at all applied volume reductions.

Bilayer thickness and lipid interface area in unilamellar extruded 1,2-diacylphosphatidylcholine liposomes: a small-angle neutron scattering study

Small-angle neutron scattering (SANS) experiments have been performed on large unilamellar liposomes prepared from 1,2-dilauroylphosphatidylcholine (DLPC), 1,2-dimyristoyl-phosphatidylcholine (DMPC) and 1,2-distearoylphosphatidylcholine (DSPC) in heavy water by extrusion through polycarbonate filters with 500 A pores. The neutron scattering intensity I(Q) in the region of scattering vectors Q corresponding to 0.0015 A(-2) < or = Q(2) < or = 0.0115 A(-2) was fitted using a step function model of bilayer neutron scattering length density and supposing that the liposomes are spherical and have a Gaussian distribution of radii. Using the lipid volumetric data, and supposing that the thickness of bilayer polar region equals to d(H) = 9+/-1 A and the water molecular volume intercalated in the bilayer polar region is the same as in the aqueous bulk aqueous phase, the steric bilayer thickness d(L), the lipid surface area A(L) and the number of water molecules per lipid molecule N intercalated in the bilayer polar region were obtained: d(L) = 41.58+/-1.93 A, A(L) = 57.18+/-1.00 A(2) and N = 6.53+/-1.93 in DLPC at 20 degrees C, d(L) = 44.26+/-1.42 A, A(L) = 60.01+/-0.75 A(2) and N = 7.37+/-1.94 in DMPC at 36 degrees C, and d(L) = 49.77+/-1.52 A, A(L) = 64.78+/-0.46 A(2) and N = 8.67+/-1.97 in DSPC at 60 degrees C. After correcting for area thermal expansivity alpha approximately 0.00417 K(-1), the lipid surface area shows a decrease with the lipid acyl chain length at 60 degrees C: A(L) = 67.56+/-1.18 A(2) in DLPC, A(L) = 66.33+/-0.83 A(2) in DMPC and A(L) = 64.78+/-0.46 A(2) in DSPC. It is also shown that a joint evaluation of SANS and small-angle X-ray scattering on unilamellar liposomes can be used to obtain the value of d(H) and the distance of the lipid phosphate group from the bilayer hydrocarbon region d(H1).

Stabilizing effect of an S-layer on liposomes towards thermal or mechanical stress

Isolated subunits of the crystalline cell surface layer (S-layer) protein of Bacillus stearothermophilus PV72/p2 were recrystallized on positively charged unilamellar liposomes. Liposomes were composed of dipalmitoylphosphatidylcholine (DPPC), cholesterol and hexadecylamine (HDA) in a molar ratio of 10:5:4 and they were prepared by the dehydration-rehydration method followed by an extrusion procedure. The S-layer protein to DPPC ratio was 5.7 nmol/micromol which approximately corresponds to the theoretical value estimated by using the areas occupied by the S-layer lattice and the lipid membrane. Coating of the positively charged liposomes with S-layer protein resulted in inversion of the zeta-potential from +29.1 mV to -27.1 mV. Covalent crosslinking of the recrystallized S-layer protein was achieved with glutaraldehyde. Chemical analysis revealed that almost all amino groups (>95%) from HDA in the liposomal membrane were involved in the reaction. To study the influence of an S-layer lattice on the stability of the liposomes, the hydrophilic marker carboxyfluoresceine (CF) was encapsulated and its release was determined for plain and S-layer-coated liposomes in the course of mechanical and thermal challenges. In comparison to plain liposomes, S-layer-coated liposomes released only half the amount of enclosed CF upon exposure to shear forces or ultrasonication as mechanical stress factors. Furthermore, temperature shifts from 25 degrees C to 55 degrees C and vice versa induced considerably less CF release from S-layer-coated than from plain liposomes. A similar stabilizing effect of the S-layer lattice was observed after glutaraldehyde treatment of plain and S-layer-coated liposomes.

Vesicle size distributions measured by flow field-flow fractionation coupled with multiangle light scattering

The separation method, flow field-flow fractionation (flow FFF), is coupled on-line with multiangle laser light scattering (MALLS) for simultaneous measurement of the size and concentration of vesicles eluting continuously from the fractionator. These size and concentration data, gathered as a function of elution time, may be used to construct both number- and mass-weighted vesicle size distributions. Unlike most competing, noninvasive methods, this flow FFF/MALLS technique enables measurement of vesicle size distributions without a separate refractive index detector, calibration using particle size standards, or prior assumptions about the shape of the size distribution. Experimentally measured size distributions of vesicles formed by extrusion and detergent removal are non-Gaussian and are fit well by the Weibull distribution. Flow FFF/MALLS reveals that both the extrusion and detergent dialysis vesicle formation methods can yield nearly size monodisperse populations with standard deviations of approximately 8% about the mean diameter. In contrast to the rather low resolution of dynamic light scattering in analyzing bimodal systems, flow FFF/MALLS is shown to resolve vesicle subpopulations that differ by much less than a factor of two in mean size.

Ultrastructural localization of B-50/growth-associated protein-43 to anterogradely transported synaptophysin-positive and calcitonin gene-related peptide-negative vesicles in the regenerating rat sciatic nerve

The growth-associated protein-43/B-50 (B-50/GAP-43) is conveyed from the neuronal soma into the axon by fast axonal transport and moved to the nerve terminal. To visualize and determine the type of vesicles by which B-50/GAP-43 is anterogradely transported in the regenerating rat sciatic nerve, we have investigated Lowicryl HM20 embedded nerve pieces dissected from the proximal side of a collection ligature. Ultrastructurally, numerous vesicular profiles of various sizes, tubules and mitochondria were seen to accumulate proximal to the collection ligature. Both, in unmyelinated and myelinated axons, B-50/GAP-43 immunoreactivity was associated with vesicular profiles which had a diameter of 50 nm. A fraction of the B-50/GAP-43 label co-localized with the small vesicle marker synaptophysin. Co-localization of B-50/GAP-43 was not detected with the large dense-core vesicle marker calcitonin gene-related peptide. These results indicate that, in rat sciatic nerve axons, B-50/GAP-43 is anterogradely transported in small 50 nm vesicles of the constitutive pathway. These transport vesicles were distinguished in two types. We suggest that one type carrying, both, B-50 GAP-43 and synaptophysin has as destination the nerve terminal, whereas the second type, which only contains B-50/GAP-43 and no synaptophysin, may be primarily targeted to the axolemma for local membrane fusion.

Mechanical properties of vesicles. II. A model for osmotic swelling and lysis

Vesicle polydispersity and leakage of solutes from the vesicle lumen influence the measurement and analysis of osmotically induced vesicle swelling and lysis, but their effects have not been considered in previous studies of these processes. In this study, a model is developed which expressly includes polydispersity and leakage effects. The companion paper demonstrated the preparation and characterization of large unilamellar lipid vesicles. A dye release technique was employed to indicate the leakage of solutes from the vesicles during osmotic swelling. Changes in vesicle size were monitored by dynamic light scattering (DLS). In explaining the results, the model identifies three stages. The first phase involves differential increases in membrane tension with strain increasing in larger vesicles before smaller ones. In the second phase, the yield point for lysis (leakage) is reached sequentially from large sizes to small sizes. In the final phase, the lumen contents and the external medium partially equilibrate under conditions of constant membrane tension. When fit to the data, the model yields information on polydispersity-corrected values for membrane area compressibility, Young's modulus, and yield point for lysis.